System and method for inputing user commands to a processor

FIELD OF THE INVENTION

p-0002This invention relates to devices configured to receive a user input.

BACKGROUND OF THE INVENTION

p-0003Entering data into a data processing device is accomplished using a data input device such as a keyboard, mouse, or joystick. Although electronic devices are constantly being miniaturized, the size of the various associated data input devices cannot be substantially decreased since they must conform to the size of the user's hands. Methods for inputting data have therefore been devised in which the user's hands do not have to touch the device. U.S. Pat. No. 5,767,842 to Korth, and U.S. Pat. No. 6,650,318 to Arnon for example, disclose an optical system in which a camera is used to monitor a user's hand and finger motions. A software application interprets these motions as operations on a physically non-existent computer keyboard or other input device. In these systems, the camera has a fixed position, so that the background of the images remains constant. This allows the software application to make use of informational present in the constant background in order to detect the user's hands in each image. This system, therefore, cannot be used in a device that in use is moved, because in this case, the background of the images is not constant, so there is no reliable background information in the images. Devices that are moved in use include hand-held devices such as a palm plot, personal digital assistant (PDA), a mobile telephone, a digital camera, and a mobile game machine.

SUMMARY OF THE INVENTION

p-0004The present invention provides a method and system for inputting operating system (OS) commands to a data processing device. The system of the invention may be used to input OS commands to the device instead of or in addition to any input devices associated with the device such as a keyboard, mouse or joystick. The system of the invention my be used in any type of data processing device such as a personal computer (PC), a portable computer such as a PDA, a laptop or a palm plot, a mobile telephone, a radio or other entertainment device, a vehicle, a digital camera, a mobile game machine, a computerized medical device and a smart house product.

p-0005The system of the invention comprises a video camera and a processor. The camera is positioned to capture images in a viewing space. Images captured by the camera are digitized by the camera and input to the processor. The processor is configured to run a software application that analyzes images captured by the camera, preferably in real time. The analysis performed by the software application includes detecting in an image obtained at a time t_ka predetermined object in the viewing space of the camera. The object may be, for example, one or more fingers of a user or a hand held stylus.

p-0006In accordance with the invention, detecting the object in an image is done using an algorithm that does not rely on background information present in the image. By not relying on background information in an image, the system of the invention may be used in devices that are moved in use. For example, the object may be detected in an image using a segmentation algorithm. Methods for segmentation are disclosed, for example, in Gonzalez, R. C. and Woods, R. E. [2002]. <i>Digital Image Processing, </i>2nd ed., Prentice Hall, Upper Saddle River, N.J.; R. Gonzalez, “Digital Image Processing using Matlab”, Pearson Prentice Hall 2004; and in Sigal, L et al “Estimation and Prediction of Evolving Color Distributions for Skin Segmentation Under Varying Illumination”, BU CS TR99-015.v2, December 1999 (revised in March 2000). Any type of segmentation algorithm may be used in the system of the invention, such as a “WaterShed” segmentation or a “growing region segmentation”, as is known in the art.

p-0007A segmentation algorithm is a multi-staged process, where at each stage the image is binarized into set of pixels having intensity (for example, either on a gray level scale or any one or more of a red, green or blue scale) below a particular threshold and the set of pixels having intensity above the threshold. At each stage, the threshold is incremented and contiguous sets of pixels in each set are identified. At the end of the process, stable sets of contiguous points are identified. A stable set of contiguous pixels is a set that remained constant during a predetermined number of consecutive stages of the process. After the stable contiguous sets of pixels have been identified in the image, an object recognition procedure is performed on the stable sets to identify the object among the stable sets. The object recognition procedure makes use of previously obtained image data relating to features of the object such as its eccentricity or dimensions. The same previously obtained image data may be applied to each image, or the image data may be updated on the basis of recent earlier images of the object. Any object recognition procedure may be applied to the stable sets. For example, an adaptive linear combination may be calculated of one or more of the features of the object and applied to each of the stable sets.

p-0008The software application also extracts from an image obtained at a time t_kone or more image analysis parameters x_i(t_k). One or more of the image analysis parameters may be history independent, such as the pixel address in the image of the tip or width of the object in pixels in the image, the length in pixels of the object in the image, or the orientation of the object in the image. One or more of the image analysis parameters may be history dependent, in which case extraction of the parameters is performed on the image together with previously and/or subsequently obtained images. Such history dependent parameters include the speed of the object at the time of the image, the change in the size of the width of the object at the time of the image, or the rate and direction of rotation of the object at the time of the image. One or more of the image analysis parameters may be binary parameters. For example, a parameter may take on the value 1 if the object is detected in the viewing space and the value 0 if the object is not detected in the viewing space. The analysis of the software thus generates a time sequence of vectors X(t_k)={x₁(t_k), . . . , x_n(t_k)} that is stored in the memory.

p-0009The software is further configured to apply one or more motion detection tests to one or more of the vectors X obtained in a recent time window and stored in the memory. The time window may be, for example 0.5 sec, or 1 sec. Each test detects a specific type of motion of the object during the time window. A test may be a binary test, in which case the result of the test is an indication 0 that the test failed (i.e. the object did not perform the motion detected by the test during the time window) or an indication 1 that the test succeeded it (i.e. the object performed the motion detected by the test). Alternatively, the result of a test may be a number between 0 and 1. One or more of the tests may be defined as a linear combination of previously defined tests. In these cases, determining whether a test succeeds or fails may be performed, for example, using a state machine, morphological algorithms, or a neural network.

p-0010As a first example, a test may be, for example, that during the time window the object approached the camera. This test succeeds if the width of the object in pixels increased during the time window. As a second example, a test may be for example that the object moved away from the camera (the width of the object in pixels decreased during the time window). Other examples of tests include that the object first approached the camera and then moved away from the camera during the time window, the object disappeared from viewing space of the camera, the object moved in a predetermined path such as in a circle or so as to trace an alphanumeric character, the object rotated, the object was stationary, the object moved (performed any type of motion), the object performed a flicking motion, the object accelerated, the object decelerated, the object moved and then stopped, or the object remained stationary at a particular location for predetermined amount of time.

p-0011The memory of the system stores a look-up table that provides, for each test an associated OS command. When a test succeeds, the OS command associated with the test is executed. The OS commands may be, for example, depressing a virtual key displayed on the display screen, moving a cursor appearing on the display screen to a new location on the screen, running on the processor a software application stored in the memory, turning on or off a light associated with the device, changing the sound level or station on a radio, or turning off the device, zooming in or out of an image displayed on a screen, controlling a surgical instrument, and sending a command to an application (eg. sending a “shooting” command to a game application).

p-0012Preferably, there is an intuitive relationship between the motion detected by a test and the OS command associated with the test. The following are examples of an intuitive relationship between a motion detected by a test and the associated OS command.

p-0013A test that detects that the object moved towards the camera and then moved away from the camera may have an associated OS command of depressing a key on a virtual keyboard to which a cursor on the screen is presently pointing.

p-0014A test that detects a linear movement of the object across the viewing space may have an associated OS command of dragging an object such as a cursor to a new location on the display screen.

p-0015A test that detects whether the object has traced out a path in the viewing space of the camera in the shape of a particular alphanumeric character may have an associated OS command of depressing a key on a virtual keyboard displayed on a display of the device for that alphanumeric character.

p-0016A test that detects a “shooting motion” of the object may have an associated OS command of “shooting” at a target appearing on the display.

p-0017A test detecting a flicking motion of the object may have an associated OS command of turning off the device.

p-0018The system of the invention may be used in devices adapted to be worn, such as a pendent or goggles. The system allows OS commands to be input to the device without touching the device. The device is thus of benefit to users whose hands are soiled, such as technicians, cooks, and surgeons. The system may also be used in devices having one or more 3-dimensional virtual objects.

p-0019Thus, in its first aspect, the invention provides a system for inputting operation system (OS) commands to a data processing device comprising:

p-0020(a) a video camera capturing images of a viewing space; and

p-0021(b) a processor configured to: <ul><li id="ul0001-0001" num="0000"><ul><li id="ul0002-0001" num="0021">i) detect a predetermined object in one or more images obtained by the camera using an object recognition algorithm not involving background information in an image;</li><li id="ul0002-0002" num="0022">ii) extract one or more image analysis parameters of the object in the one or more images obtained by the camera; and</li><li id="ul0002-0003" num="0023">iii) for each of one or more motion detection tests: <ul><li id="ul0003-0001" num="0024">(I) applying the motion detection test to image analysis parameters extracted during a recent time window; and</li><li id="ul0003-0002" num="0025">(II) executing an operating system command associated with the motion detection test if the motion detection test succeeds.</li></ul></li></ul></li></ul>

p-0022In its second aspect, the invention provides a data processing device comprising the system for inputting operation system (OS) commands of the invention.

p-0023In its third aspect, the invention provides a method for inputting operation system (OS) commands to a data processing device having a video camera capturing images of a viewing space, comprising: <ul><li id="ul0004-0001" num="0000"><ul><li id="ul0005-0001" num="0028">a. detecting a predetermined object in one or more images obtained by the camera using an object recognition algorithm not involving background information of an image;</li><li id="ul0005-0002" num="0029">b. extracting one or more image analysis parameters of the object in the one or more images obtained by the camera; and</li><li id="ul0005-0003" num="0030">c. for each of one or more motion detection tests: <ul><li id="ul0006-0001" num="0031">i. applying the motion detection test to image analysis parameters extracted during a recent time window; and</li><li id="ul0006-0002" num="0032">ii. executing an operating system command associated with the motion detection test if the motion detection test succeeds.</li></ul></li></ul></li></ul>

p-0024In its fourth aspect, the invention provides a computer program comprising computer program code means for performing all the steps of the method of the invention.

p-0025In its fifth aspect, the invention provides a computer program of the invention embodied on a computer readable medium.

BRIEF DESCRIPTION OF THE DRAWINGS

p-0026In order to understand the invention and to see how it may be carried out in practice, a preferred embodiment will now be described, by way of non-limiting example only, with reference to the accompanying drawings, in which:

p-0027<figref idrefs="DRAWINGS">FIG. 1</figref> shows a device having a system for inputting OS commands to a data processing device in accordance with one embodiment of the invention; and

p-0028<figref idrefs="DRAWINGS">FIG. 2</figref> shows a method for inputting OS commands to a data processing device in accordance with the method of the invention.

DETAILED DESCRIPTION OF THE INVENTION

p-0029<figref idrefs="DRAWINGS">FIG. 1</figref> shows a data processing device <b>2</b> in accordance with one embodiment of the invention. The data processing device <b>2</b> may be, for example, a personal computer (PC), a portable computer such as a PDA, a laptop or a palm plot, or a mobile telephone, a radio or other entertainment device, a vehicle, a digital camera or a mobile game machine. The device <b>2</b> has a video camera <b>6</b>. The device <b>2</b> may also be provided with a display screen <b>4</b> and various data input devices such as a keypad <b>8</b> having a plurality of keys <b>10</b> for inputting data into the data input device <b>2</b>.

p-0030The device <b>2</b> has a processor <b>12</b> inside housing <b>14</b>. The processor <b>12</b> is configured to receive data from the camera <b>6</b> and any other input devices associated with the device <b>2</b> such as the keypad <b>8</b>.

p-0031The camera <b>6</b> views a conical or pyramidal volume of space <b>16</b> indicated by the broken lines. The camera <b>6</b> may have a fixed position on the device <b>2</b>, in which case the viewing space <b>16</b> is fixed relative to the device <b>2</b>, or may be positionable on the device <b>2</b>, in which case the viewing space <b>16</b> is selectable relative to the device <b>2</b>. Images captured by the camera <b>6</b> are digitized by the camera <b>6</b> and input to the processor <b>12</b>.

p-0032The processor <b>12</b> is configured to run a software application that analyzes images captured by the camera <b>6</b>. The analysis performed by the software application includes detecting in an image obtained at a time t_ka predetermined object <b>24</b> in the viewing space <b>16</b>. As explained above the object <b>24</b> is detected using an object recognition algorithm that does not involve background information of the image, for example, by using a segmentation algorithm. The object <b>24</b> may be for example a finger of a user as shown in <figref idrefs="DRAWINGS">FIG. 2</figref>. Alternatively, the object may be a hand held stylus (not shown). The object <b>24</b> has a length and orientation in the viewing space represented by means of a vector <b>27</b> having a tip <b>25</b> at a free end of the object <b>24</b>. The object <b>24</b> also has a width <b>23</b>. The user may use his other hand <b>19</b> to hold the device <b>2</b> in use, if the device <b>2</b> is a hand-held device. The hand <b>19</b> may also be used to activate real input devices associated with the device <b>2</b>, such as activating keys <b>10</b> on the keypad <b>8</b>.

p-0033The software application is further configured to extract from an image obtained at a time t_kone or more image analysis parameters x_i(t_k). One or more of the image analysis parameters may be history independent, such as the pixel address in the image of the tip <b>25</b>, the width <b>23</b> of the object <b>24</b> in pixels in the image, the length in pixels of the object <b>24</b> in the image, or the angle that the vector <b>27</b> makes in the image with a fixed direction. One or more of the image analysis parameters may be history dependent, in which case extraction of the parameters is performed on the image together with previously and/or subsequently obtained images. Such history dependent parameters include the speed of the tip <b>25</b> at the time t_kof the image, the change in the size of the width <b>23</b> of the object <b>24</b> at the time of the image, or the rate of rotation of the vector <b>27</b> at the time of the image. One or more of the image analysis parameters may be binary parameters. For example, a parameter may take on the value 1 if the object <b>24</b> is detected in the viewing space and the value 0 if the object <b>24</b> is not detected in the viewing space. The analysis of the processor thus generates a time sequence of vectors X_k={x₁(t_k), . . . , x_n(t_k)} that is stored in the memory <b>16</b>.

p-0034The software application also applies one or more motion detection tests to one or more of the vectors X_kobtained in a recent time window and stored in the memory <b>16</b>. The time window may be, for example 0.5 sec, or 1 sec. The length of the time window may be different for different tests. Each test detects a specific type of motion of the object <b>24</b> during the time window. The result of each test is an indication that the test failed (i.e. the object <b>24</b> did not perform the motion detected by the test during the time window) or an indication that the test succeeded it (i.e. the object <b>24</b> performed the motion detected by the test). As explained above, a test may or may not be a binary test. The result of test may depend upon a user input. For example, a test may give an indication of success only if the user simultaneously depressed a predetermined key on the keypad <b>8</b>.

p-0035As a first example, a test may be, for example, that during the time window the object <b>24</b> approached the camera. This test produces an indication of success if the width <b>23</b> of the object in pixels increased during the time window. As a second example, a test may be for example that the object moved away from the camera (the width <b>23</b> of the object in pixels decreased during the time window). Other examples of tests include that the object <b>24</b> first approached the camera and then moved away from the camera during the time window, the object disappeared from viewing space <b>16</b>, the object moved in a predetermined path, the object rotated, the object was stationary, the object moved (performed any type of motion), the object performed a flicking motion, the object accelerated, the object decelerated, or the object moved and then stopped.

p-0036The memory <b>16</b> stores a look-up table that provides, for each test an associated OS command. When a test succeeds, the OS command associated with the test is executed. The OS commands may be, for example, depressing a virtual key displayed on the display screen, moving a cursor appearing on the display screen to a new location on the screen, running on the processor <b>12</b> a software application stored in the memory <b>16</b>, or turning off the device <b>2</b>. The device may provide an indication that the OS command was executed. For example, an OS command equivalent to depressing a key on the virtual keyboard may be indicated by briefly showing the key depressed on a virtual keyboard on the screen <b>4</b>, or by briefly changing the appearance of the key. Other possibilities for indicating that the OS command was executed include briefly enlarging or otherwise changing the appearance of a depressed key or of the cursor on the screen <b>4</b>, displaying an icon on the screen <b>4</b>, producing a sound, and vibrating the device.

p-0037The viewing space <b>16</b> may be divided into two or more independent subspaces, with different tests being applied to motion of the object <b>24</b> in the different subspaces.

p-0038<figref idrefs="DRAWINGS">FIG. 2</figref> shows a flow chart for a method for inputting an OS command to a data processing device in accordance with one embodiment of the method of the invention. In step <b>30</b> an image obtained by the camera <b>6</b> is input to the memory <b>16</b> of the processor <b>12</b>. In step <b>31</b>, the predetermined object <b>24</b> is identified in the image using an object recognition procedure that does not involve background information in the image. In step <b>32</b>, one or more image analysis parameters are extracted from the image and stored in the memory <b>16</b>. In step <b>34</b> a motion detection test is applied to the extracted image analysis parameters obtained during a recent time window and stored in the memory <b>16</b>. In step <b>36</b> the indication of the test is determined. If the indication is “success”, i.e. the test detected the type of motion detected by the test, then in step <b>38</b>, the OS command associated with the test is looked up in the look-up table stored in the memory <b>16</b>, and in step <b>40</b>, the OS command associated with the test is executed. If at step <b>36</b> it is determined that the indication of the test is “failure”, i.e. the test did not detect the type of motion detected by the test, or after step <b>40</b>, if the indication of the test was 1, the process proceeds to step <b>42</b> where it is determined whether another motion test is to be applied to the image. If yes, the process returns to step <b>34</b> with another motion detection test being applied to the image. If no (i.e. all of the motion detection tests have been applied to the present time window) then the process returns to step <b>20</b> with a new image obtained by the camera <b>6</b> being input to the memory <b>16</b>.

p-0039It will also be understood that the system according to the invention may be a suitably programmed computer. Likewise, the invention contemplates a computer program being readable by a computer for executing the method of the invention. The invention further contemplates a machine-readable memory tangibly embodying a program of instructions executable by the machine for executing the method of the invention.